The following abbreviations are herewith defined:
3GPP third generation partnership project
ACK acknowledgement
CAZAC constant amplitude zero auto-correlation
CDM code division multiplex
CQI channel quality indication
DFT discrete Fourier transform
DM demodulation
e- evolved (also known as LTE for e-UTRAN)
FDM/FDMA frequency division multiplex/multiple access
IFFT inverse fast Fourier transform
LB long block
LTE long term evolution (also known as 3.9G)
NACK negative ACK
Node B base station or BS (including e-Node B)
OFDM orthogonal frequency division mutiplex
PUCCH physical uplink control channel
RAN radio access network
RLC radio link control
RS reference signal
RU resource unit
SIMO single input multiple output
TTI transmission time interval
UE user equipment
UL uplink
UMTS universal mobile telecommunications system
UTRAN UMTS terrestrial radio access network
V-MIMO virtual multiple input/multiple output
ZC Zadoff-Chu
Reference can be made to 3GPP TR 25.814, V7.0.0 (2006-06); TECHNICAL SPECIFICATION GROUP RADIO ACCESS NETWORK; PHYSICAL LAYER ASPECTS FOR EVOLVED UNIVERSAL TERRESTRIAL RADIO ACCESS (UTRA) (Release 7), such as generally in section 9.1 for a description of the SC-FDMA UL of e-UTRA. Referring to FIG. 1A, which reproduces FIG. 9.1.1-4 of 3GPP TR 25.814; according to that (former) format there are two blocks reserved for the pilot signal in the 3GPP LTE UL former frame format, referred to as short blocks SB1 and SB2. That format has been recently changed, and FIG. 1B shows a generic format according to current adoption, taken from section 4.1, FIG. 1 of 3GPP TS 36.211 (V1.0.0) (2007-03). It is seen at FIG. 1B that according to the current format, there are no longer SBs but rather the structure is one subframe consisting of two slots, each of length 0.5 msec. The SBs of the older format are replaced by LBs in the newer format. Regardless of the particular format though (FIGS. 1A, 1B or otherwise), in each subframe there will be two pilots (2 pilot LBs in the latest format or more generically two pilot RSs). Additional LBs may also be used for this purpose (e.g., for transmitting out-band or out-time RSs), which may or may not be periodic.
More specifically, as is described in Section 9.1 of 3GPP TR 25.814, the basic uplink transmission scheme is single-carrier transmission (SC-FDMA) with cyclic prefix to achieve uplink inter-user orthogonality and to enable efficient frequency-domain equalization at the receiver side. Frequency-domain generation of the signal, sometimes known as DFT-spread OFDM (DFT S-OFDM), is assumed and illustrated in FIG. 1C, which reproduces FIG. 9.1.1-1 of 3GPP TR 25.814. This approach allows for a relatively high degree of commonality with the downlink OFDM scheme and the same parameters, e.g., clock frequency, can be reused.
The basic sub-frame structure formerly approved for the UL transmission is shown herein in FIG. 1A; two short blocks (SB) and six long blocks (LB) are defined per sub-frame, and two subframes span one TTI. Short blocks are used for reference signals for coherent demodulation and/or control/data transmission. Long blocks are used for control and/or data transmission. As seen at FIG. 1B, there is no longer a distinction as between SBs and LBs but there are still two slots, each to bear one pilot sequence. The data could include either or both of scheduled data transmission and non-scheduled data transmission, and the same sub-frame structure is used for both localized and distributed transmission.
The Zadoff-Chu CAZAC sequence has been agreed upon as the pilot sequence for the LTE UL.
ZC sequences and their modified versions (i.e., truncated and/or extended ZC sequences) are therefore used as reference signals in the LTE uplink system, and will also be used on the physical uplink control channel (PUCCH). It has been decided in 3GPP that data-non-associated control signals such as ACK/NACK and CQI will be transmitted on PUCCH by means of ZC sequences. A paper entitled “MULTIPLEXING OF L1/L2 CONTROL SIGNALS BETWEEN UEs IN THE ABSENCE OF UL DATA” (3GPP TSG RAN WG1 Meeting #47bis, Sorrento, Italy; Jan. 15-19, 2007 by Nokia, document R1-070394) is a reference for those methods. Multiple UEs in a given cell share the same Zadoff-Chu sequence while keeping the orthogonality by using a cyclic shift specific to each UE. In this manner different ones of the UEs in a cell may multiplex their UL transmissions (e.g., non-data associated UL transmissions) on the same frequency and time resource (physical resource block/unit or PRB/PRU; currently 180 kHz in LTE). The orthogonality of the ZC sequences enables the receiving Node B to discern the different signals from one another. However, two problems arise.
First, ZC sequences of different lengths may occasionally have large cross correlation properties. This causes an interference problem for demodulation reference signals.
In order to avoid “code-domain” collisions on PUCCH, different cells/sectors should utilize different ZC mother sequences. This is a problem related to ZC sequences used in PUCCH in that there are not enough proper mother sequences for sufficient randomization, so in some instances adjacent cells operate with the same ZC mother sequence (sometimes termed the base sequence).
Another issue related to PUCCH is that different UEs transmitting data-non-associated control signals in the same cell are separated only by means of different cyclic shifts of the same ZC sequence. The problem with this approach is that the sequences are not perfectly orthogonal against each other.                Orthogonality is Doppler-limited with block-wise spreading performed in the time domain; and        Orthogonality is delay-spread-limited when using cyclic shifts of ZC or CAZAC codes within a LB.        
It is also noted that orthogonality problems will increase when some practical limitations such as power control errors are taken into account.
FIG. 2 is a schematic diagram showing the available cyclic shifts for a ZC sequence of length 12 symbols. It is noted that orthogonality between different code channels varies widely; the best orthogonality is achieved between the code channels which have the largest difference in cyclic shift domain (e.g., cyclic shift #0 and cyclic shift #6 of FIG. 2) whereas the worst orthogonality is between two adjacent cyclic shifts (e.g., cyclic shift #3 and cyclic shifts #2 and #4 of FIG. 2).
The same issue is related also to the cyclic shifts of block-level spreading codes (see the above-referenced document R1-070394 for further details). Considering an extreme case where the Doppler spread is very high (i.e., due to the UE movement). It is noted that block level codes with adjacent cyclic shifts have the worst cross-correlation properties, and are therefore most difficult to distinguish from one another at the receiver after being multiplexed. Further detail as to addressing such Doppler shifts can be seen at U.S. Provisional Patent Application No. 60/899,861, filed on Feb. 5, 2007; and now PCT/IB2007/004134, filed on Dec. 28, 2007.
Pseudo-random cyclic shift hopping is known in the art, as can be seen at a paper entitled: “CYCLIC SHIFT HOPPING FOR UPLINK SOUNDING REFERENCE SIGNAL” (3GPP TSG RAN WG1 Meeting #48, St. Louis, USA, Feb. 12-16, 2007 by ETRI, document R1-070748).
Another relevant paper is entitled “NON-COHERENT ACK/NACK SIGNALING USING CODE SEQUENCES AS INDICATORS IN E-UTRAN UPLINK” (3GPP TSG RAN WG1 Meeting #47 bis, Sorrento, Italy, Jan. 15-19, 2007 by ETRI, document R1-070078). This paper proposes to use some kind of randomization for ACK/NACK signaling. It assumes that the ACK/NACK signal is transmitted without separate RS such that a certain cyclic shift of CAZAC code corresponds to an ACK and another cyclic shift corresponds to a NACK, respectively. Document R1-070078 appears to propose that mapping of ACK/NACK is done such that a one-to-one mapping relation between the ACK/NACK information and the transmission cyclic shifts in the second block is reversed against the mapping in the first long block LB, and the ACK/NACK information is conveyed in the amount of the cyclic shift.
This is seen to forego what the inventors see as the primary advantage of cyclic shifting: randomizing interference between different code channels when the same underlying mother ZC sequence is used. Where the cyclic shift is given by the ACK/NACK message the UE seeks to send, the orthogonality of the ZC codes cannot be maximized. As will be seen below, the inventors have devised a different approach to address the problem of too few ZC mother codes available to orthogonalize all ZC sequences in use by the various UEs.